Topological Data Analysis for Classification of DeepSat-4 Dataset
Published in 2020 10th International Symposium onTelecommunications (IST) IEEE, 2020
Recommended citation: Moghadam, Mehdi Hosseini, and Mir Mohsen Pedram. "Topological Data Analysis for Classification of DeepSat-4 Dataset." In 2020 10th International Symposium onTelecommunications (IST), pp. 246-250. IEEE, 2020. https://ieeexplore.ieee.org/abstract/document/9345829/
Topological Data Analysis (TDA) is a new emerging and fast growing field of data science providing a set of tools from algebra, topology, and geometry to extract features from data based on its topological and geometrical features. This paper combines available methods from topological data analysis including persistent homology, persistent entropy, and persistent diagrams to build a strong topological feature extractor model from the topological properties of an image. By feeding features extracted by the topological model to machine learning models, we perform the classification task on DeepSat (SAT-4) dataset.
Recommended citation: Moghadam, Mehdi Hosseini, and Mir Mohsen Pedram. “Topological Data Analysis for Classification of DeepSat-4 Dataset.” In 2020 10th International Symposium onTelecommunications (IST), pp. 246-250. IEEE, 2020.